Marina Michalak, Mariusz Tszydel*, Products of Caddis-fly Larvae Trichoptera( ) Jadwiga Bilska, Izabella Krucińska Silk Glands as a New Natural Textile Fibre Technical University of Łódź, Abstract Department of Textile Marketing and Engineering, Among the invertebrates, some are able to produce natural silk fibres which can ul. Żeromskiego 116, 90-543 Łódź, Poland serve as an alternative to the threads of the spider’s web. The larvae of the caddis-fly [email protected] make use of a multifunctional silky floss in producing hunting nets and residential tubular structures. Many species of caddis-fly build their capture webs in fast-flowing rivers. *Department of Ecology and Vertebrate Zoology, Therefore we suggest that such fibres must be strong. However, these webs have been University of Łódź, studied and researched previously, taking into consideration the webs’ shape and the ul. Banacha 12/16, 90-237 Łódź, Poland geometry of the meshes. This paper presents preliminary results concerning some of the [email protected] web material’s morphological properties and tensile strength. We investigated the product of silk gland coming from Hydropsyche pellucidula, which is one of the most common in Polish lowland rivers. Our researches proved that a cross-section of caddis- fly fibres is circular, with an average diameter of 10 µm. The mean tensile strength of the fibres investigated is 45 cN/tex (5.7⋅109 N/m2), and is comparable to the strength of most spider thread. Key words: natural silk fibers, silk gland, morphological properties, caddis-flies, tensile strength, Hydropsyche pellucidula.

Caddis-fly larvae are similar to spiders, and detritus (unidentified, disintegrated butterflycaterpillars (e.g. silkworms) organic matter) can fall into it [19]. and blackflies Simuliidae ( , Diptera) in their ability to produce a cocoon shell in Each species has its own web mesh the form of a multifunctional silky floss. shape and diameter, and the way the web Taking into account the variety of forms is mounted depends on environmental and versatility of silken products, it may conditions (Figure 3) [10]. be concluded the mastery of making co- coons is comparable in Trichoptera and Irrespective of hydrological conditions, Araneinae. webs are usually highly regular in their n Introduction mesh size and pattern (Figure 4). Insects are the most widely distributed The fact that the caddis-fly’s ability to The structure and shape of a web may and frequently the most abundant group spin silky floss was known in the past, also depend on temperature [6]. The of invertebrates in the world. They in- not only to biologolists, is confirmed by minimum temperature at which webs clude caddis-flies (Trichoptera), which the text: are built is also a characteristic feature are not a commonly known species. They “O what tangled webs the caddis weave, are merolimnic insects, whose larvae and when first they practise to deceive!” pupae, with few exceptions, prefer aquat- ‘Marmion, Sir Walter Scott, 1808. ic ecosystems, while the adults (imagos) a) are pharate stages. The larvae of caddis- Larvae of caddis-flies use a silky floss flies are found in all types of freshwater which is a product of their spinning and marine environments. glands (Figure 1) to build various ac- commodation structures. The ‘weaving’ is done by the elements of the mouthpart b) and not by the abdomen, as in the case of spiders, which makes it much more Figure 2. Simple structures in the form of precise. holes and hunting tunnels of caddis-flies: a) Lype reducta [9], b) Wormaldia occipi-

RESEARCH ANDRESEARCH DEVELOPMENT – CURRENT PUBLICATIONS talis [16]. Some caddis-fly larvae build only simple structures: padded holes, corridors, hunt- ing tunnels (Figure 2). The larvae can a) easily move inside them. Such holes may b) have continuations in the form of ‘signal- ling threads’, which inform the predatory caddis-flies about the prey that stumbles a) over the threads, or about intruders [5]. Others build fully-fledged webs. These b) have a shape of a tunnel-like sieve, inside which the caddis-fly awaits its prey. The Figure 1. Location of spinning glands: inlet of such a structure is always directed a) in the body of the caddis-fly Neureclipsis Figure 3. Web shape of Hydropsyche bimaculata, b) the opening of silk glands at against the water current, so not only live angustipennis, depending on the substratum: the haustellum [2]. organisms but also fragments of plants a) plant [13], b) stone [19].

28 FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54) FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54) 29 build their webs in river currents. They a) are usually stretched between aquatic b) plants, mainly submerged macrophytes or floating leaves (Figure 6).

The most spectacular builder in this fam- ily is Plectrocnemia conspersa. We know [26] at least three types of webs built by c) P. conspersa, depending on water depth and velocity. In rivers of fast current, the web resembles a –‘swallow’s nest’ length of mesh width of mesh (Figure 7a), and a ‘hammock’ in shallow streams (up to 5 cm deep) (Figure 7b). In d) area of mesh deeper rivers the web is placed on the bot- tom, with the funnel-like openings turned Figure 6. Different types of webs and web- Figure 4. Web of caddis-fly larvae of the upwards, in this way catching victims houses of caddis-fly larvae from the family genus Hydropsyche [20]. moving along the bottom (invertebrates) : a) Holocentropus or settling organic matter (Figure 7c). dubius, b) Cyrnus flavidus, c) and d) N. bimaculata. a and b [23], c [13], d [2]. Using the silky floss to build a web is not this fabric’s only application. Without the floss, it would be impossible to form another characteristic structure, the house. a) Usually, the larva begins building its house by spinning a silky tubular structure, into which it builds elements forming the house (Figure 8): fragments of plants, detritus, sand, small stones and so on [14, 21, 15].

A strong floss and an organic or inorganic substrate are also the basic components b) of the house built for the pupa (Figure 9 - see page 30). All types of caddis-flies, in the fifth larval stage, start building this, because without the cocoon it would be im- c) possible for the specimen to metamorphose into the later stage. Figure 7. Different types of webs made by Caddis-fly larvae usually build their webs P. conspersa depending on environmental conditions in running waters: a. fast-mov- in fast-flowing rivers, and so the cocoon ing river, b. shallow river, c. deep river [5]. shell they produce must be strong.

The product of Trichoptera’s spinning (silk) glands could be an alternative to a Figure 5. Scheme of web building by spider’s floss. Unlike spiders, caddis-flies H. angustipennis [19]. fairly often produce cocoon shells in man- made conditions. It is much easier to breed for a given species (in low temperatures them, because cannibalism among them

RESEARCH ANDRESEARCH DEVELOPMENT – CURRENT PUBLICATIONS some Hydropsychidae species, e.g. is not observed, which does happen in H. angustipennis, do not build webs) spiders (Arachnida). Unfortunately, we do [18]. However, web building always con- not find much significant data in literature sists of a set of similar actions, which are on the properties of fibres made by - cad shown in Figure 5. dis-flies. Therefore the aim of the studies that have been undertaken is to analyse One of the most impressive webs, tubu- the physical-mechanical properties, com- lar in shape, is built by N. bimaculata position and chemical structure of the (Polycentropodidae); it may be 20 cm cocoon shell which these insects produce. long, whereas the larva itself is no longer than 2 cm. Despite being so big, its mesh n is incredibly small, 0.45×0.45 µm; this is Study material one of the denser webs built by caddis- For microscopy observations, samples Figure 8. The course of the initial phase of flies [17]. Larvae of the family Polycen- were selected for the present studies from house-building whose construction requires tropodidae, with a few exceptions, do not different types of cocoon shell made by the product of spinning glands [16].

28 FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54) FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54) 29 sary to determine the linear density of the fibres studied. In order to do this, after determining the breaking force and elon- gation, the broken parts of the fibres were placed on a micro-slide in an alcoholic medium, and then under the microscope; the crosswise dimensions of fibres were determined. In this case a Lanametr pro- jection microscope was used. The value of the crosswise dimension as determined by an optical method was considered on the fibre diameter. The mean diameter dav was determined for each fibre on the basis of 20 measurements. Later, its linear Figure 9. Pupal case and the structure of floss weave in the cocoon of larva Rhyacophila density Tt was calculated. subovata [13]. n Results Table 1. Tensile properties of Caddisfly fibres. Microscopy observations No. of F, cN ε, % σ, cN/tex L, cN cm d , µm Tt, tex sample av Images of fibres taken from the studied 1 10.31 23.8 26.2 2.45 25 0.393 samples are presented in Figures 10a - 2 5.40 16.0 21.5 0.86 20 0.251 10g. They were registered on the stand 3 5.40 18.0 21.5 0.97 20 0.251 for computer image analysis. The figures 4 5.45 22.5 16.0 1.23 22 0.304 show the lengthwise views of the fibres. 5 4.41 14.6 19.4 0.64 19 0.227 All the images were obtained at the same 6 2.06 6.8 22.9 0.14 12 0.090 magnification. In order to make the 7 5.40 18.0 21.5 0.97 20 0.251 visual analysis of the images easier, a 8 7.95 21.5 23.9 1.71 23 0.332 micrometric grid with a scale, obtained at 9 3.38 11.2 13.5 0.38 20 0.251 the same magnification at which the pho- 10 4.75 14.4 44.8 0.68 13 0.106 tographs presented in Figures 10a-10g 11 5.53 11.9 24.5 0.66 19 0.226 were taken, is presented on each figure. 12 4.21 10.7 12.7 0.45 23 0.332 13 4.35 9.6 17.3 0.42 20 0.251 The images obtained are insufficiently accurate to enable clear determination of the shape of the fibre’s cross-section. Add- larvae of Hydropsyche pellucidula (Tri- Samples for microscopic studies were itional investigations made with a scan- choptera, Hydropsychidae). These cad- prepared in the form of permanent mi- ning electron microscope (SEM) showed dis-flies inhabit the lowland River Drze- croscopy slides (placed on micro-slides) that the fibres’ cross-sections were circu- wiczka (the biggest right-bank tributary in the environment in which they were lar. An example of the results obtained is of the River Pilica). The material studied stored; for studies of tensile strength, the presented in Figure 11. In some cases two included webs, structural floss forming a samples were taken at random only from fibres were connected together, as we can construction skeleton of shelters for case- structural floss stored in ethyl alcohol. see in the weave structures presented in less larvae, and floss from cocoon shell. Figures 10a and 10f, and as a double fibre Some samples were taken directly from Tensile strength tests cross-section in Figure 12. In such a case, the river and preserved in 70% ethanol. Tensile strength tests were carried out us- the proportion of the largest to the small- Other samples originated from labora- ing an Instron tensile tester. The distance est crosswise dimensions can be more tory-bred insects and were stored in dis- between clamps was 10 mm, and the than 2:1, and the fibres therefore appear tilled water. The tensile test were carried elongation speed 10 mm/min. The length to be flat (Figure 10f). out on only one type of fibre material, on of the fibres did not enable them to be the structural flosses from cocoon shells. clamped directly in the tensile tester, so Tensile strength tests before measurements they were placed in The test results are given in Table 1. The n Methodology of studies paper frames and stored in alcohol. Meas- table contains values of breaking force F urements of tensile strength were carried for each sample studied, expressed in cN, Microscopy observations out just after they were removed from values of relative elongation ε, in%, read Fibres selected from the studied samples the alcohol. In the laboratory, climate at the peak of the curve of the drawing, were observed using computer image conditions were normal (t = 20 °C ± 2%, the crosswise dimension of each fibre analysis, on a measuring stand composed Rh = 65% ± 5%). For every sample we subjected to breaking dav, expressed in of a microscope (Biolar), a CCD Elemis determined the values of breaking force, µm. The values of thickness are mean KK 35 video camera and a PC computer. of absolute elongation and relative elon- values of 20 measurements for each fibre. Lucia software was used, enabling the gation at break. On the basis of breaking dav values are given in µm. The samples automatic separation of objects from the force measurements, the tenacity was cal- were numbered according to the order of background [12]. culated. To make calculations, it is neces- measurements.

30 FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54) FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54) 31 strength is 37.26%, and of elongation at a) b) break 34.4%. Additionaly, the work to break (in cN cm) was calculated for eve- ry sample and also presented in Table 1. It is clearly visible that high elongation at break is decisive for a great work needed to break the fibre.

n Discussion This work presents preliminary results obtained for the structural floss of only c) d) one type of cocoon shell.

The preliminary results show that the specific strength of fibres produced by caddis-flies is very high, reaching 45 cN/tex (5.7×109 N/m2). The value of specific strength estimated on the basis of the results obtained for Trichoptera fibres is 1.5 times higher than the specific strength of steel [26], and is also as high as the strength of spider’s floss, which is only two times higher than the strength of e) f) steel [27]. However, the specific strength of Trichoptera fibres is four times lower than the specific strength of aramid fi- bres [7] - (180 - 190 cN/tex). According to [3], caddis-fly silk is one of the weakest g) natural silks so far reported, with a mean tensile strength of (221±22)×106 N/m2. Our results (5.7×109 N/m2) show that the product of the trichoptera salivary gland reaches the breaking strength and elastic Figure 10. Microscopic views of weave structure and flosses; a) weave structure in the web modulus of spider dragline silk (up to with places marked where single fibres are knotted, b) weave structure in the cocoon shell, 2×109 N/m2 and 30×109 N/m2 [25-11] and c) floss from a cocoon shell, d) structural floss forming a construction skeleton of the shelter, e) lengthwise projection of a floss taken from a cocoon shell, f) structural floss forming a highly drawn nylon (0.7×109 N m2 and construction skeleton of the shelter, g) lengthwise projection of structural (skeleton) floss. 2.4×109 N/m2 [1]. The tensile strength of caddis fibres is comparable to the same parameter of the threads of a spider’s web. However, if we compare spider dragline silk to Kevlar, we find that the silk has lower strength [8].

If further studies confirm these pre- liminary results, it will be possible to use caddis-fly fibres for different technical purposes, especially for use in aquatic conditions, because natural caddis floss is made and used in water. It is quite possible that they can also be used as Figure 11. Fibre cross-section of a structural Figure 12. The cross-section of a double (skeleton) floss. fibre of the structural (skeleton) floss. surgical threads or bio-bandages, if they turn out to be biocompatible with human tissues, especially as nature created them Assuming that d is the fibre diameter, On the basis of these results, the specific av to be used in an aquatic environment. the linear density Tt of each fibre stud- strength of each fibre and the mean value Studies are currently under way on the ied was calculated. The density of fibres of specific strength were calculated. Val- chemical composition of caddis floss and was assumed to be the value of density ues of tenacity range from 12.7 cN/tex to on its biocompatibility with tissues of of ethyl alcohol, that is 800 kg/m3. This 44.8 cN/tex. The mean value of tenac- live organisms. value was assumed on the base of observ- ity is 22.1 cN/tex, and the mean relative ing the behaviour of the studied fibres in elongation at break is 15.3%. There is a Studies conducted in [24] on the cocoon ethyl alcohol. The linear density values great dispersion of measurement results. shells of Rhyacophila fasciata demon- are given in tex. The variation coefficient of specific strated the exceptional properties of this

30 FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54) FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54) 31 material. This regular matter was formed Acknowledgment 15. Nepomnyashchikh V. A., 1992. The Con- from floss 5 µm in diameter and a mesh trol of Selection of Particles for Case of 30×40 µm, and it has the properties of We would like to express our gratitude to Building by Larvae of Chaetopteryx Vi- Maria Grzybkowska and Janusz Majecki for a semi-permeable membrane, selective losa Fabr. (Trichoptera, Limnephilidae). their comments and help in preparing biolo- In: Proc. 7th Int. Symp. On Trichoptera, only to some ions: CO 2-, Cl-, NO3-, 3 gical material, as well as Henryk Wrzosek for (Ed.): C. Otto, Bachhuys Pub., Umea, 2- + + 2+ 2+ SO4 , N , K , Mg or Ca . Thus there taking microscope pictures. 263-265. are chances that this structure could be 16. Nilsen A., 1948. Postembryonic Develop- used as intelligent clothing. Each species ment and Biology of the Hydroptilidae. has its own characteristic parameters of A Contribution to the Phylogeny of the floss diameter and mesh density. There- References Caddis-flies and to the Question of the Origin of the Case-Building Instinct. Det. fore they can be used for selections of 1. Billmeyer, F. W. (1984). Textbook of Poly- Kgl. Danske Vidensk. Selskab, Biol. ions. Additionally, it should be taken into mer Science N. Y., John Wiley & Sons. Skrifter, 5: 1-200. account that caddis floss is sometimes as 2. Brickenstein C., 1955. Über den Netzbau 17. Petersen R. C. Jr., Petersen B-M., µ der Larve von L.. fine as spider floss (starting from 0.5 m Wallace J. B., 1984. Influence of Velo- Abhandl. Bayer. Acad. Wiss. Mathem. in the case of floss from the cocoon shell city and Food Availability on Catchnet - Naturwiss. Klasse, München, N. F., of Limnephilus vittatus [4]). This is of Dimensions of Neureclipsis Bimaculata 69: 1-44. (Trichoptera, Polycentropodidae). Holorc. crucial importance in various measuring 3. Brown S. A., Ruxton G. D., Humphries S. optical devices. Now we can be even (2004). “Physical Properties of Hydropsy- Ecol., 7: 380-389. more precise, thanks to the Trichoptera’s che Siltalai (Trichoptera) Net Silk”. Jour- 18. Philipson G. N., Moorhouse B. H., 1974. cocoon shell. Certainly, production nal Of The North American Benthological Observations on Ventilatory and Net- Spinning Activities of The Larvae of the costs will decide whether caddis floss Society: Vol. 23, No. 4, Pp. 771–779. Genus Hydropsyche Pict. (Trichoptera, can be successfully competitive with 4. Cianficconi F., Bicchierai M. C., Moretti G., 1992. Silk Glands And Silk Weave Hydropsychidae) Under Experimental nanofibres, whose technology is being In Trichopteran Larvae. In: Proc. 7th Int. Conditions. Freshwat. Biol., 4: 525-533. worked on. Symp. On Trichoptera, (Ed.): C. Otto, 19. Sattler W., 1958. Beitrage zur Kenntnis Bachhuys Pub., Umea, 33-38. von Lebensweise und Körperbau der n 5. Edington J. M., Hildrew A. G., 1995. A Re- Larve und Puppe von Hydropsyche Pict. Conclusions vised Key To The Caseless Caddis Larvae (Trichoptera) mit besonderer Berücksich- 1. The investigations presented were in Of The British Isles With Notes On Their tigung des Netzbaues. Zeitschr. Morph. the nature of a ‘reconnaissance’, and Ecology. Freshwat. Biol. Assoc., 1-135. Ökol. Tiere 47, 2: 115-192. 6. Fey J. M., Schumacher H., 1978. Zum did not encompass all the possible 20. Tachet H., Pierrot J. P., Bournaud M., Einfluss Wechselnder Temperatur auf 1987. Distribution of the Hydropsyche physical and chemical properties of den Netzbau von Larven der Köcher- Larva And The Structure of Their Nets. caddis threads. Therefore our data fliegen - Art. Hydropsyche Pellucidula In: Proc. 5th Int. Symp. on Trichoptera, does not finish such research. (Trichoptera, Hydropsychidae). Entomol. eds. M. Bournaud & H. Tachet, Lyon, 2. The tensile strength of Trichoptera Ger., 4: 1-11. France, 293-297. salivary glands does not differ sig- 7. Gessner W., 1992. Neue Möglichkeiten 21. Tomaszewski C., 1981. The Principles of mit neuen Aramidfasern. Technische nificantly from the strength of spider Case Building Behavior in Trichoptera. Textilien/Technical Textiles, 35: 128-129. In: Proc. 3rd Int. Symp. on Trichoptera, thread or anthropogenic fibres such 8. Gosline, J. M., M. E. Demont, et al. ed. G.P. Moretti, Junk, The Hague, as nylon or aramid fibres. This data is (1986). "The Structure And Properties Of 365-373. comparable. Spider Silk." Endeavour 10(1): 37-43. 22. Townsend C. R., Hildrew A. G., 1979. 9. Hickin N. E., 1946. Larvae Of The British 3. The small diameter of thread and reg- Form And Function Of The Prey Catching Trichoptera. Trans. R. Entomol. Soc., ular, dense meshes in pupal cocoons Net Of Plectrocnemia Conspersa Curt. London, 97: 187-212. function as a semi-permeable mem- Larvae (Trichoptera, Polycentropodidae). 10. Hildrew A. G., Edington J. M., 1979. Oikos 33: 412-418. brane, which gives medicine hopes Factors Facilitating The Coexistence of devising an intelligent bandage on Of Hydropsychid Caddis Larvae In The 23. Wesenberg-Lund C., 1911. Biologische their basis. Same River System. J. Anim. Ecol., 48: Studien über den Netzspinnenden Tri- 4. So far, the special ‘sister’ structure 557-576. chopteren Larven. Int. Rev. Gesamten Hydrobiol. (Biol. Suppl.) 3: 1-64. (visible on the cross-section) of cad- 11. Kaplan, D. L., S. J. Lombardi, et al. (1991, 24. Wichard W., Schmidt H. H., Wagner R., dis fibres are unsubstantiated. It may In Press). ‘Silks: Chemistry, Properties And Genetics.’ In: Biomaterials: Novel Ma- 1992. The Semi-Permeability Of The Pu- yield exceptional properties. This as- terials From Biological Sources; D. Byrom pal Cocoon Of Rhyacophila (Trichoptera: pect will be investigated in subsequent (Ed.). New York, Stockton Press. 3-53. Spicipalpia). In: Proc. 7th Int. Symp. On research. 12. Krucińska I., Zastosowanie Komputerowej Trichoptera, ed. C. Otto, Bachhuys Pub., 5. Further investigation into the structure Analizy Obrazu Do Oceny Budowy Run Umeĺ, 33-38. and properties of fibres produced by Włóknistych. PAN, Łódź, 2000. 25. Zemlin, J. C. (1968). ‘A Study of the Me- caddis-fly larvae should enlarge the 13. Lepneva S. G., 1964. Fauna ZSSR. Chru- chanical Behavior of Spider Silks.’ Natick ściki Tom II, Cz.1, Larwy I Poczwarki An- Ma, U.S. Army Natick Laboratories. Re- current state of knowledge of this nulipalpia. Wydawnictwo Nauka, 1-562. port # 69-29-Cm (Ad 684333). kind of fibres. However, the test mate- 14. Lepneva S. G., 1971. Fauna of The 26. http://www.dupont.com/kevlar/ rial should be considerably larger and USSR. Trichoptera 1, Larvae and Pupae whatiskevlar.html. more differentiated. of . Translation from 1964 27. http://pajaki.zwierzeta.eurocity.pl. Russian edition. Jerusalem, Israel Pro- gram For Scientific Translations, 1-638. Received 05.04.2005 Reviewed 24.10.2005

32 FIBRES & TEXTILES in Eastern Europe January / December 2005, Vol. 13, No. 6 (54)